In December 2011 Twitter unveiled a new UI, along with updated iPhone, Android, and web clients. The response from tech circles was immediate and extremely negative. Twitter 4.0 for iPhone got slammed particularly hard. Here’s John Gruber with a pretty good representation of the views expressed by many:

Twitter 4.0 for iPhone lacks the surprise, delight, and attention to detail of a deserving successor to Tweetie, offering instead a least common denominator experience that no one deserves.

By the time Twitter 4.0 came out I was already using Tweetbot, but I updated and played around with the app anyway, as I’m sure many did. I had three main issues:

• Severely limited functionality. Some things I do all the time in Tweetbot are either impossible or very difficult to do in Twitter for iPhone. This includes easily switching accounts, adding/removing people from lists, seeing someone’s @-reply stream, and quickly getting to saved searches.
• Inability to interact with tweets in the main stream. You can’t click on a link or someone’s profile from the main Tweet stream. You have to tap through to the Tweet detail — in many ways an unnecessary tap. More on this later.
• The Discover tab. Like most complainers I assumed Discover was just a thinly veiled attempt to start shoving ads in our faces. Back in Twitter 4.0 this was just a list of seemingly random tweets, probably based on some global trending topics. There was always talk of customization, but the initial incarnation of Discover didn’t have much of that.

So, like many others, I joked about it on Twitter1, and then moved back to Tweetbot without another thought.

Why didn’t they just come clean and call the “Discover” tab the “Monetization” tab? #NewNewTwitter

— Rian van der Merwe (@RianVDM) December 9, 2011

But that was not the end of the story. Slowly but surely, Twitter has been working on putting the pieces together of that consistent user experience they’ve been talking about for a long time.

The story unfolds

In June 2012, Twitter rolled out expanded Tweets, a way to see more information about a single Tweet (like an embedded photo or an article summary). They called the technology behind this feature Twitter Cards.

Then, in July, Twitter 4.3 for iPhone came out with support for expanded Tweets. This was followed by Twitter 5.0 in September, which included a redesigned iPad app (a topic for a different blog post, so let’s just leave that for now), as well as profile header images.

And then came Twitter 5.1 on November 16. It was a point release, sure, but I think this is the version that finally brings together two separate threads that Twitter has been working on for a while: Twitter Cards and the Discover tab. The release notes for Twitter 5.1 said this:

See what’s popular within your network on Discover.

— Tweets, tailored just for you, now appear right in the stream

— These Tweets show photo, video, and article previews so you can engage easily

This time something weird happened in my Twitter stream. I started seeing a few positive tweets about the new app. I even saw a few positive mentions about the Discover tab. Fred Wilson blogged about it just today.

I decided to take another look. I worked with Alex to implement Twitter cards on this site. I moved Tweetbot to another screen and committed to trying Twitter for iPhone as my primary app for a week. And now I think I finally see where Twitter is going with all of this. And that maybe we should have trusted them a year ago. Possibly even apologize to them. But I’m getting ahead of myself. Let’s back up.

Information consumption on Twitter

When it comes to information consumption on Twitter, I think there are two overriding user needs:

• Get through as much content in as little time as possible.
• Know as quickly as possible if a link is worth clicking on.

The first requirement is technically difficult, but conceptually easy to meet. Just make the app as fast as it possibly can be. The second requirement is more difficult though. In the context of Twitter (specifically on mobile) there are two pieces of information that is important to decide if a link is worth clicking on:

• The source. This is easy to tell if you can see the URL, but since so many people still use URL-shorteners like bitly, the domain is often obscured, so you don’t know where the link is going to end up.
• A summary of the article. This is not easy to do in 120 characters2, especially if the Tweet just says “This!! —> bitly.com/blahblah”.

Why not just click on a link to see if it’s worth reading? Well, because it messes with that speed principle. Loading a site takes time. Especially if you don’t know if the bitly link goes to Mashable and you then have to download a 2MB page with a gazillion HTTP requests. Clicking on a link is expensive, so you only want to do it if it’s worth it.

So this is where Twitter Cards come in. If someone tweets a link to sites that have implemented it, you can immediately see the source and a summary of the article to help you figure out if it’s really worth clicking through — even if a URL shortener is being used:

You also have to tap through to the tweet detail to click on the link; you can’t do it from the main stream. As I mentioned earlier, this extra tap used to annoy me, but I now think it’s a deliberate and important design decision. They are compromising immediate convenience for clarity of information. You might not agree with the decision, but it’s certainly not an oversight — I’m pretty sure it’s well thought through.

The implementation reminds me of the distinction between search results pages and product details pages on e-commerce sites. A search for “The Beatles” on Amazon doesn’t show you an “Add to cart” button on the search results page. You have to go to the product details page for that:

Twitter’s approach is similar. You have to view the summary before you can “commit”. The goal is to keep you inside the app until you’re absolutely sure you’re ready for the “purchase” — which in this case means clicking on a URL, emailing it to someone, sending it to Instapaper, etc. And now that the Discover tab is a much better customized version of photos and articles you might be interested in, the entire story is coming together really nicely.

There are still many features I miss in Twitter for iPhone. Having to poke around aimlessly for a while every time I’d like to see Tweets from a different list is probably the most frustrating issue at this point. But I have to say that Twitter Cards have made my experience so much more enjoyable and efficient that I’m going to stick with Twitter 5.1 for iPhone beyond my one-week experiment.

So what have we learned?

There are also some important product lessons to learn here. In his brilliant essay Subcompact Publishing Craig Mod sums it up nicely:

In product design, the simplest thought exercise is to make additions. It’s the easiest way to make an Old Thing feel like a New Thing. The more difficult exercise is to reconsider the product in the context of now. A now which may be very different from the then in which the product was originally conceived.

This is exactly what Twitter has done, starting with 4.0 almost a year ago. It was a gutsy move to rethink the entire experience — one they got a lot of ridicule for. And I’m sure the design team felt like this quite often:

(Source: The Oatmeal)

But to their credit, they stuck to their guns. They knew where they were going, and instead of surrendering to the extremely vocal complainers, they kept their eyes on the vision and went for it. That’s good product management. And now, almost a year later, I think they are finally seeing the tide turning as we’re getting a better sense of the end game.3

So, uh, I believe an apology is in order. I’m sorry, Twitter. I see where you’re going with this. And I like it.

---

1. Oh, the irony. ↩

2. Not a typo. You need to leave 20 characters for the URL, ok? ↩

3. Yes, I know, there’s the API debacle as well, but I’m talking specifically about the Product design aspect here. ↩

Meagan Fallone wrote a great article on social entrepreneurship for Fast Company. From Technology Is Useless If It Doesn’t Address A Human Need:

We in turn can teach Silicon Valley about the human link between the design function and the impact for a human being’s quality of life. We do not regard the users of technology as “customers,” but as human beings whose lives must be improved by the demystification of and access to technology. Otherwise, technology has no place in the basic human needs we see in the developing world. Sustainable design of technology must address real challenges; this is non-negotiable for us. Social enterprise stands alone in its responsibility to ensuring sustainability and impact in every possible aspect of our work.

There is much we can learn from this approach. Even in the consumer space, we need to replace some “customer” thinking with “human” thinking and look for ways to improve people’s lives, not just get more money from them.

In another great article on social entrepreneurship, David Bornstein quotes Sally Osberg, president and chief executive of the Skoll Foundation:

“I’ve come to see how the ‘social’ that characterizes their purpose also characterizes their way of working. In other words, social entrepreneurs don’t just pursue a social end, they pursue that end in a fundamentally communal way.” This approach is badly needed at a time of extreme factionalism, she adds: “Regardless of whether you call it teamwork, collaboration or consensus-building, we need it, and we need it now.”

I’ve seen this first-hand in our work with Praekelt Foundation. Their passion for their work, clarity of purpose, and relentless pursuit of working together to create the best possible experience, is teaching me so much about how powerful design can be — in any context.

Phillip Kruger argues that responsive design shouldn’t be used in Africa in a Memeburn article called Why Google might just be wrong about responsive design in Africa. He lays out his argument in two parts. First:

Responsive design only works on smartphones. So by default you are already ignoring 80+% of users in Africa. You are also reaching the 20% of users that possibly have internet access at home or work.

This is an argument I see a lot, but it’s valid only in the context of target audience and use cases. If the target market for your service primarily uses low-end phones, then by all means don’t bother with responsive design. But let’s say you’re building a site to order take-out food and deliver in major cities, the situation changes dramatically. Now you’re most likely looking at a target market that sits squarely in the 20% of people who have smartphone access (and who don’t want to get off their couches and walk to their PCs to place an order).

This is why personas, scenarios, and use cases are so important. If you’re building a service for ALL THE PEOPLE (which isn’t advisable), then averages are appropriate. And those averages will rightly guide you to focusing on the 80% of people who do not have smartphones. But in most cases, the analytics that matter are not the averages of all users, but the specifics of the market you’re going after. Don’t dismiss responsive design in Africa because of averages. Dismiss it if it doesn’t make sense for your target market.

Phillip continues:

Responsive design is not lightweight. When using responsive design, the size of the download to the browser is still very big (in fact it’s very similar to what the webpage would be). All the HTML is still being downloaded (even parts that are hidden on a smartphone if you use media queries to set display:none in your CSS). Sure, you can have rules to download separate images for separate display sizes and that should help a bit.

Tim Lind wrote a comment on the article that addresses the problem with this logic:

Responsive design does not mean you can’t do server side optimisations. In fact it can help you to do these, and encourages a more efficient design process. You can design the content for the feature phones, and responsive media queries allows you to upgrade the design with a single stylesheet file for the smartphone or desktop (which server side optimisation could exclude).

One of the many good things about responsive design is that it forces designers and developers to spend a lot of time on optimisation to ensure light, fast pages. This is just good practice for web development in general — not just for mobile. Page bloat is a huge problem, with the average web page now being more than 1 MB big.

Page speed optimisation is just good web citizenship, and it should be a requirement regardless of whether or not a responsive design approach is taken. The other point to remember is that mobile networks already do a lot of compression on served images (see How should we handle responsive images?).

What worries me about this debate is that there appears to be no room for nuance. Responsive design is either the answer to all of Africa’s problems, or we shouldn’t do it at all. But as with most things, the appropriate approach is to say “it depends.” A mobile strategy shouldn’t be a decision between a native app or a separate mobile site. A mobile strategy should form part of a larger web strategy, and it needs to include a discussion about the appropriateness of responsive design. It might not be the right thing for your project, but it should be on the table.

I keep reminding myself of Ben Callahan’s statement in The Responsive Dip: “Just because you can’t, doesn’t mean you shouldn’t.” Just because this is a difficult problem that we haven’t quite figured out, doesn’t mean we should throw it away and go back to how we’ve always done things. What we need to do now is push through and find elegant ways to apply responsive design in Africa. Where it makes sense, of course.

Update 2012/11/26: Phillip responded to all the feedback on his Memeburn post. See Google might be wrong - part 2. It’s good to get additional clarification on the Google talk that formed the backdrop for his original post. This isn’t the last discussion we’ll have about RWD in Africa, and that’s a good thing. We need to figure this out…

There’s quite a fight going on in the comments of Elliot Jay Stocks’s A conversation with Erik Spiekermann. If you’re able to wade through the mudslinging you’ll find some good points in there, like this paragraph by Erik himself:

Design is first and foremost an intellectual activity which has nothing to do with what medium you work in. It is about looking at a problem, understanding it, translating it into visuals, actions, and messages. That is solving the problem, whatever medium the solution may end up in.

The worst work is done by designers who have decided on a medium before they even know the problem that has to be solved. Just like a print designer (and I do not make that distinction myself) should not immediately think brochure or poster, an interaction designer should also be able to think about other media besides websites or apps. Otherwise you end up behaving like the infamous hammer: every problem looks like a nail.

This relates to a point I made earlier that we shouldn’t let technology or devices (what Erik calls “the medium”) guide product decisions. The problem and the use cases should guide those decisions.

Dmitry Fadeyev makes the best case so far for why it’s not a good idea to combine mobile and desktop operating systems into a unified experience, like Windows 8 has done. From Blurring of the Lines:

The road to a good OS is not a blurring of the lines between PCs and tablets, but rather an amplification of the differences through a strong focus on the uses that each category serves. The desktop OS should make use of large screen real estate and the precise targeting of the mouse cursor. The mobile OS should be optimized for the small screen and for the rough tap of the finger. The desktop OS should focus on power users and multi-tasking, the mobile OS should focus on content consumption. The environments they run on are different, the use cases are different, and the solutions should be different.

That’s exactly right. This “unified experience” sounds like a decision made from the viewpoint of devices and technology, not use cases. For example, if you make decisions based on devices and technology, you may decide to create an iPhone app before you know what kind of phones people will use your service on. If you make decisions based on real use cases, you may actually find that very few people would use your service on a mobile device, so a better solution would be to (gasp!) optimise for desktop use1.

The irony is that even though Microsoft made a huge deal about their “no compromise” design philosophy, the Windows 8 experience will have to make compromises if the same software needs to work on both mobile and desktop devices.

---

1. Wait, don’t slaughter me. I love Mobile First. I’m just saying that some services or applications just don’t lend themselves to mobile use. I’d argue that tax return software falls into that category. ↩

In The Banner Blindness Cure: How Fewer Ads Can Equal More Revenue Dave Zinman points out something all readers know already, and publishers will hopefully take note of:

It’s no wonder, looking at these stats, that banner blindness is such a glaring issue. Talk about losing sight of the forest for the trees: We’re so busy looking after our bottom line, we’re not paying attention to the user experience. We hit our visitors over the heads with ads like sledgehammers, then wonder why our ads aren’t “performing.” It’s absurd. Clearly, we’re doing it wrong. […]

Wouldn’t a publisher be far better off serving fewer ads, and taking top dollar for one or two premium placements? With highly relevant ads that aren’t forced to compete against several other ads on the page, odds of interaction and possible conversion are tremendously improved. And when ads perform better, publishers, advertisers and consumers win.

Somewhat related, here’s what’s happening over on the far end of the creepiness scale:

The odds are that access to you — or at least the online you — is being bought and sold in less than the blink of an eye. On the Web, powerful algorithms are sizing you up, based on myriad data points: what you Google, the sites you visit, the ads you click. Then, in real time, the chance to show you an ad is auctioned to the highest bidder.

Not that you’d know it. These days in the hyperkinetic world of digital advertising, all of this happens automatically, and imperceptibly, to most consumers.

If I may use a John Gruberism: Gross.

The Atlantic printed an interesting interview with Clay Shirky, covering a wide range of topics like privacy, publishing, and the Internet as a distractor. Shirky argues for tempered pessimism about the oft-lamented distracting role of the Internet. Here’s why:

The other case for tempered pessimism is that the examples we have of group creation don’t rely on wholesale change — whether you are looking at examples of amateur collaboration (digitizing old ship logs, figuring out how proteins fold), sites of cultural production (Pinterest, YouTube), collaborative consumption (Freecycle, CouchSurfing) or new kinds of conversational value (Quora, Reddit). Each of these initiatives requires only a small percentage of the population to donate a small percentage of time to making or sharing to have an outsized effect.

This is, for me, the biggest driving force in our use of the cognitive surplus: considering that by the end of the 20th century, the total time spent in media consumption, with no accompanying production or sharing and even precious little annotation or discussion, is a situation so different from ours in the early 21st century.

His point is that even though we’re much more connected to media (which definitely has its drawbacks), it’s a much less passive connection than it used to be. Now we comment, like, share, and in the best case scenario, further discussions in a meaningful way. And that’s a good thing.

Stephen Hackett’s Parenting Technology is a haunting piece of writing for The Magazine. I don’t want to give the story away, so I’ll just quote a couple of key paragraphs:

How many little moments have I missed in my kids’ lives by checking Twitter on my iPhone while they play in the yard? How many hours have I spent writing or hacking away on my Web site while I should have been reading books to Josiah?

Technology saved my son’s life, yet has left him with terrible scars. It allows me to work from the hospital on days when I need to, but distracts me from being engaged at home. Technology is wonderful, but terrible, all at once.

How we deal with that balance — with what technology wants — remains one of the biggest struggles of our time.

Christopher Butler shares a long and interesting agency perspective on responsive design in What We’ve Learned About Responsive Design. He shares tips, case studies, and some unresolved issues — including how to deal with larger display sizes:

So, the parting question for me is this: What about upward responsiveness? If we’re heading toward bigger displays with much higher pixel density, how will our designs adapt to make use of them? We’re all excited and sold on the concept of responsive design, but so far that has been limited to responding to smaller conditions. If we’re up for that challenge, than I know we can do better on the “desktop,” too.

So far we’ve dealt with that issue either by setting a maximum width for websites, or to live with acres of white space. Christopher rightly points out that we need to experiment more with ways to take advantage of larger displays without overwhelming users with too much content.

(link via @smashingmag)

Dan Frommer believes that when it comes to live events, Twitter Stands Alone:

Then look over at Twitter, where the room is bursting with fresh news, links, photos from everywhere, alerts that Karl Rove is melting down or that Diane Sawyer seems wasted, jokes coming so fast that you can barely keep up. (Many of them even funny.) You control the content, the sources, the volume, the pace, and your drink. Sometimes, it’s wrong, but it’s quickly corrected, and you should be more skeptical anyway. And if you want, you can participate. You’re not just watching.

This sentiment resonates with me — especially because we don’t have cable TV at home. On election night (morning in South Africa) I went to the gym at 5:30am so I could watch CNN on the TVs there (two birds with one stone and all that). But all the TVs in the section I was in when the race was called were set to sports channels, so how did I find out who won? Twitter, of course.

Twitter is also my first port of call when there is an issue on this site, or with one of the apps I use. If my site is down, the first thing I do after running a traceroute is send a tweet to Cloudflare or Mediatemple. When Tweetbot isn’t sending push notifications, I just search for “Tweetbot notifications” to find out if it’s just me. This has been said before, but there is simply no service out there that is better for real-time search.

Ok, that was more of a Twitter love letter than I thought it would be. I’ll stop now.